Re: [PATCH V2 06/15] perf tools: fix parse_events_terms() freeinglocal variable on error path

From: Adrian Hunter
Date: Fri Jun 28 2013 - 02:26:45 EST


On 27/06/13 19:13, David Ahern wrote:
> On 6/27/13 1:54 AM, Adrian Hunter wrote:
>> The list_head is on the stack, so just free the rest of the list.
>>
>> Signed-off-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
>> ---
>> tools/perf/util/parse-events.c | 7 ++++++-
>> tools/perf/util/parse-events.h | 1 +
>> tools/perf/util/pmu.c | 2 +-
>> 3 files changed, 8 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
>> index 995fc25..d9cb055 100644
>> --- a/tools/perf/util/parse-events.c
>> +++ b/tools/perf/util/parse-events.c
>> @@ -1231,12 +1231,17 @@ int parse_events_term__clone(struct
>> parse_events_term **new,
>> term->val.str, term->val.num);
>> }
>>
>> -void parse_events__free_terms(struct list_head *terms)
>> +void parse_events__free_terms_only(struct list_head *terms)
>> {
>> struct parse_events_term *term, *h;
>>
>> list_for_each_entry_safe(term, h, terms, list)
>> free(term);
>> +}
>> +
>> +void parse_events__free_terms(struct list_head *terms)
>> +{
>> + parse_events__free_terms_only(terms);
>>
>> free(terms);
>> }
>
> Seems like you can just fix parse_events__free_terms to do the right thing
> and not call free(terms) -- ie., no need for a separate "only" function.
> From there only test_term() needs to do the free(terms) and since it mallocs
> the memory there it is nicely symmetrical.

There are 2 more calls to parse_events__free_terms in util/parse-events.y

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/